Transcript The History of the EEG - uni

The ElectroEncephaloGramm
Cognitive Neuropsychology
January 16th, 2001
Outline
1.
2.
3.
4.
5.
History of the EEG
Biological Foundations of the EEG
Measuring the EEG
Analyzing the EEG
Applications of the EEG
The History of the EEG
1875 Caton records brain potentials from cortex
1883 Marxow discovers evoked potentials
1929 Berger records electrical activity from the
skull
1936 Gray Walter finds abnormal activity with
tumors
1957 The toposcope (imaging of electrical
brain activity)
1980 Color brain mapping (quantitative EEG)
Hans Berger – EEG Pioneer
In 1929, Hans Berger
• Recorded brain activity from the closed
skull
• Reportet brain activity changes
according to the functional state of the
brain
– Sleep
– Hypnothesis
– Pathological states (epilepsy)
First EEG recorded by Berger
Gray Walter – Brain Imaging
The toposcope by Gray Walter
In 1957, Gray Walter
• Makes recordings with
large numbers of
electrodes
• Visualizes brain activity
with the toposcope
• Shows that brain
rhythms change
according to the mental
task demanded
Outline
1. History of the EEG
2. Biological Foundations of the EEG
1. Brain Rhythms
2. Information Processing in the Neocortex
3. Summation Potentials
3. Measuring the EEG
4. Analyzing the EEG
5. Applications of the EEG
EEG in the States of Vigilance
Frequency Ranges
Beta:
Alpha:
Theta:
Delta:
14 – 30 Hz
8 – 13 Hz
5 – 7 Hz
1 – 4 Hz
Alpha Rhythm
Frequency:
Amplitude:
Location:
State of Mind:
8 – 13 Hz
5 – 100 microVolt
Occipital, Parietal
Alert Restfulness
Alpha blockade occurs when new stimulus is processed
Source: oscillating thalamic pacemaker neurons
Beta Rhythm
Frequency:
Amplitude:
Location:
State of Mind:
14 – 30 Hz
2 – 20 microVolt
Frontal
Mental Activity
Reflects specific information processing between cortex
and thalamus
Theta Rhythm
Frequency:
Amplitude:
Location:
State of Mind:
5 – 7 Hz
5 – 100 microVolt
Frontal, Temporal
Sleepiness
Nucleus reticularis slows oscillating thalamic neurons
Therefore diminished sensory throughput to cortex
Delta Rhythm
Frequency:
Amplitude:
Location:
State of Mind:
1 – 4 Hz
20 – 200 microVolt
Variable
Deep sleep
Oscillations in Thalamus and deep cortical layers
Usually inibited by ARAS (Ascending Reticular
Activation System)
Outline
1. History of the EEG
2. Biological Foundations of the EEG
1. Brain Rhythms
2. Information Processing in the Neocortex
3. Summation Potentials
3. Measuring the EEG
4. Analyzing the EEG
5. Applications of the EEG
Cortex Structure
The neocortex consists of six
distinct layers
I Molecular layer
II External granular layer
III External pyramidal layer
IV Internal granular layer
V Internal pyramidal layer
VI Polymorphic or
multiform layer
Cortex Structure Layer I
I Molecular layer
Molekularschicht
• Apical dendrites of
pyramidal cells
• Axons of stellate cells
(parallel to cortex surface)
• Few cell bodies
• Local (intracortical)
information exchange
Cortex Structure Layer II/ III
II External granular layer
Äußere Körnerschicht
III External pyramidal layer
Äußere Pyramidenschicht
• Stellate & Small Pyramidal
Cells
• Intercortical Information
Exchange
– Afferent fibers from other
cortical areals enter the layers
– Association (to the same
hemisphere) and commisural
(to the other hemisphere) fibers
leave cortex (reentry at
destination)
Cortex Structure Layer IV
IV Internal granular layer
Innere Körnerschicht
• Stellate cells
• Afferents from Thalamus
• Numerous and complex
synaptic connections
• Relay of thalamic
information to other
cortical layers
• Information input layer
(well developed in sensory
cortex)
Cortex Structure Layer V
V Internal pyramidal layer
Innere Pyramidenschicht
• Large pyramidal cells
• Projection fibers to
subthalamic brain areas
– Basal ganglia
– Brain stem
– Spinal chord
• Information output layer
(well developed in motor
cortex)
Cortex Structure Layer VI
VI Multiform layer
Spindelzellschicht
• Neurons of various shapes
(mainly fusiform)
• Adjacent to white matter
• Corticothalamical
information exchang
Cytoarchitecture Neocortex
Input layers:
II/IV (granular)
Output layers: III/V (pyramidal)
1: heterotypical agranular
cortex
– Mainly pyramidal layers
(output)
– Primary motor cortex
5: heterotypical granular
cortex
– Mainly granular layers
(input)
– Primary sensory cortex
2-4: homotypical cortex
– Association areas
– All layers developed
Outline
1. History of the EEG
2. Biological Foundations of the EEG
1. Brain Rhythms
2. Information Processing in the Neocortex
3. Summation Potentials
3. Measuring the EEG
4. Analyzing the EEG
5. Applications of the EEG
Summation Potentials
The EEG measures
• not action potentials
• not summation of
action potentials
• but summation of
graded Post
Synaptic Potentials
(PSPs)
(only pyramidal cells:
dipoles between
soma and apical
dendrites)
Outline
1. History of the EEG
2. Biological Foundations of the EEG
3. Measuring the EEG –
The international 10/20 system
4. Analyzing the EEG
5. Applications of the EEG
The International 10/20
System
Terminology: 10/20 System
Nasion:
Inion:
Location:
Numbers:
point between the forehead and the skull
bump at the back of the skull
Frontal, Temporal, Parietal, Occipital, Central
z for the central line
Even numbers (2,4,6) right hemisphere, odd (1,3,5) left
EEG channels
Channel: Recording from a pair of electrodes (here with a
common reference: A1 – left ear)
Multichannel EEG recording: up to 40 channels recorded in
parallel
Participants with Electrodes
EEG in clinical diagnostics
EEG in scientific research
Outline
1.
2.
3.
4.
History of the EEG
Biological Foundations of the EEG
Measuring the EEG
Analyzing the EEG
1. Event Related Potentials
2. Spectral Analysis
3. Topographical Mapping
5. Applications of the EEG
Event Related Potentials
• Averaging of trials
following a stimulus
• Noise reduction: The
noise decreases by the
squareroot of the number
of trials
• Far field potentials require
up to 1000 measurements
• Assumption: no
habituation occurs
(participants don‘t get
used to stimulation)
Language specific ERP
Components
• N400: Semantic mismatch
marker
• P600: Syntactic mismatch
marker
Example Sentences:
Correct (Baseline): The cats won't eat
the food Mary gives them.
Semantic mismatch: The cats won't
bake the food Mary gives them.
Syntactic mismatch: The cats won't
eating the food Mary gives them.
Semantic and syntactic mismatch: The
cats won't baking the food Mary gives
them.
EEG Spectral Analysis
• Fast Fourier Transform seperates spontaneous EEG
signal to component frequencies and amplitudes
• Restriction: high frequency resolution demands long
(in the range of seconds) analysis windows
Topographical Maps
Topographical maps plot EEG data on a map of the
brain.
Data is interpolated between electrodes.
Usual data plotted:
• ERP maps
– potential changes
• Spectral maps
– frequency changes
• Statistical maps
– comparison of
measurements
Outline
1.
2.
3.
4.
5.
History of the EEG
Biological Foundations of the EEG
Measuring the EEG
Analyzing the EEG
Applications of the EEG
Weiss, Rappelsberger (2000)
Long-range EEG synchronization during word
encoding correlates with successful memory
performance
Methods Section 1
A set of 19 gold-cup electrodes was
glued to the scalp according to the
international
10/20-placement
system. Data were recorded against
the average signals of both earlobes
((A1 + A2) /2) which turned out to be
the most suitable reference for
coherence
analysis.
The
electrooculogram
(EOG)
was
recorded from two electrodes located
at the left later outer cantus and above
the right eye. Electrode impedance did
not exceed 8 kΩ and signal bandpass
was 0.3 –35 Hz. Data were
simultaneously monitored by an inkwriter system and digitally sampled at
256 Hz to be stored on hard disk.
After recording, the EEG data were
screened for artefacts (eye blinks,
horizontal and vertical eye movements,
muscle activities) by visual inspection
on a monitor and on paper. These two
methods allowed a very reliable
exclusion of the artefacts. Impedance
did not exceed 8 kΩ and signal
bandpass was 0.3 –35 Hz. Data were
simultaneously monitored by an inkwriter system and digitally sampled at
256 Hz to be stored on hard disk. After
recording, the EEG data were
screened for artefacts (eye blinks,
horizontal and vertical eye movements,
muscle activities) by visual inspection
on a monitor and on paper. These two
methods allowed a very reliable
exclusion of the artefacts.
Methods Section 2
EEG
was
recorded
during
memorization of the different lists
of nouns and during four
interspersed resting periods with
eyes open lasting one minute
each. According to the behavioral
results epochs of recalled and of
not recalled ones were selected for
further analysis. The beginning of
each noun was marked by a trigger
and the following 1 s EEG epoch
was Fourier-transformed. All 1-s
artefact-free epochs of the resting
EEG
were
also
Fouriertransformed. On the average, per
subject, 16  4 epochs for recalled
nouns auditorily presented were
analysed, 28  5 for not recalled
nouns auditorily presented, 7  2
for
recalled
nuons
visually
presented, 14  4 for not recalled
nouns visually presented and 198
 45 for the resting EEG. Then
averaged power spectra and
cross-power
spectra
were
computed for each subject.
According to the 19 elctrode
positions, 19 averaged power
spectra were computed. Cross
power spectra were computed.
Cross
power
spectra
were
computed between all possible
pairs, which yielded 171 values per
frequency.
Methods Section 3
To reduce the large data set the
adjacent spectral values vere
averaged to obtain broadband
parameters for the following
frequency bands: delta-1 (1 – 2
Hz), delta-2 (3 – 4 Hz), theta (5 – 7
Hz), alpha-1 (8 – 10 Hz), alpha-2
(11 – 12 Hz), and beta-1 (13 – 18
Hz). Finally, 19 mean amplitudes
(square root of power) per
frequency band were computed
and the normalization of the 171
cross-power spectra yielded 171
coherence values per frequency
band. Grand mean values were
obtained by averaging amplitude
and coherence values across
subjects.
Since it has been demonstrated
that,
especially,
lower
EEG
frequencies were correlated with
memory
processes,
we
predominantely investigated lower
frequency bands in the present
study. The division into distinct,
well-selected frequency bands was
made since several studies point at
their different functional role during
cognitive processing.
Coherence Map
A coherence map plots differences in
coherence between recalled and not recalled
nouns.
Results
• Overall increase of coherence for recalled vs.
not recalled nouns
• Long range synchronization of frontal and
temporal/parietal neuronal assemblies
increases for recalled nouns.
Outline
1.
2.
3.
4.
5.
History of the EEG
Biological Foundations of the EEG
Measuring the EEG
Analyzing the EEG
Applications of the EEG
Thank you for your attention!